Fabry disease is an X-linked inborn error of metabolism resulting from a deficiency of the lysosomal enzyme, .alpha.-galactosidase A. Deficiency of .alpha.-galactosidase A results in the accumulation of its major glycosphingolipid substrate, globotriaosylceramide and related glycolipids with terminal .alpha.-galactosidic linkages. Progressive substrate deposition, especially in the plasma and vascular endothelium, leads to ischemia and infarction with early demise due to vascular disease of the heart and kidney.
Fabry diseases is one of approximately 30 lysosomal storage diseases known to affect humans. Each of these diseases result from an inherited trait which affects the levels of enzymes in the lyosome. Tay-Sach's disease and Gauucher disease are members of this group of diseases. Since specific pathways for the uptake of these other lysosomal enzymes also exist, enzyme replacement therapy is expected to be effective in Fabry disease and could logically be expected to be successful in these other diseases as well. Although these diseases are individually rare, (e.g., several thousand patients with Fabry disease are known to occur world wide, i.e., 1 to 40,000), as a group this class of diseases accounts for a significant fraction of all inherited diseases.
Several lines of evidence suggest that enzyme replacement therapy may be beneficial for patients with Fabry disease. For example, it has been demonstrated in cell cultures of fibroblasts obtained from patients with this disease that enzyme present in the culture medium is specifically transported to lysosomes. Clinical trials of enzyme replacement therapy have been reported for patients with Fabry disease using infusions of normal plasma (Mapes et al., 1970, Science 169: 987-989); .alpha.-galactosidase A purified from placenta (Brady et al., 1973, New Eng. J. Med. 279: 1163); or .alpha.-galactosidase A purified from spleen or plasma (Desnick et al., 1979, Proc. Natl. Acad. Sci. USA 76: 5326-5330). In one study (Desnick et al.) intravenous injection of purified enzyme resulted in a transient reduction in the plasma levels of the substrate, globtriaosylceramide. However, due to the limited availability of the human enzyme obtained from human sources, insufficient quantities were available for further study.
The .alpha.-galactosidase A enzyme is a lysosomal enzyme which hydrolyzes globotriaosylceramide and related glycolipids which have terminal .alpha.-galactosidase linkages. It is a 45 kDa N-glycosylated protein encoded on the long arm of the X chromosome. The initial glycosylated forms (Mr=55,000 to 58,000) synthesized in human fibroblasts or Chang liver cells are processed to a mature glycosylated form (Mr=50,000). The mature active enzyme as purified from human tissues and plasma is a homodimer. (Bishop et al., 1986, Proc. Natl. Acad. Sci. USA 83: 4859-4863).
A human liver cDNA for .alpha.-galactosidase A was identified in a .lambda.gt11 expression library (Calhoun et al., 1985, Proc. Natl. Acad. Sci. USA 82: 7364-7368), and its sequence reported by Bishop et al. The original cDNA isolated by Calhoun et al. encoded the mature amino acid sequence of .alpha.-galactosidase A but did not contain the complete signal peptide sequence of the precursor form.
The partial cDNA clone was used to construct an E. coli expression vector by placing the .alpha.-galactosidase A coding sequence under control of the trp promoter (Hantzopoulos et al., 1987, Gene 57: 159-169). The level of expression of active enzyme was sufficient to support E. coli. growth using an .alpha.-galactoside substrate as the sole carbon source; however, increased levels of the 45 kDa protein were not detected in Coomassie blue-stained gels upon trp promoter induction nor was biological activity detected in vitro.
A genomic clone was later isolated which carried the promoter and first exon of the protein including the full signal peptide (Quinn et al., 1987, Gene 58: 177-188). Clones of full length cDNAs of a precursor .alpha.-galactosidase A from human fibroblasts were reported (Tsuji et al., 1987, Eur. J. Biochem. 165: 275-280) and used to obtain transient expression of the enzyme in monkey COS cells. The level of enzyme activity reported was only about 40% above background.
Hence one long standing need of the prior art is to provide large quantities of active human .alpha.-galactosidase A, especially for use in enzyme replacement therapy. To achieve this goal, a full length cDNA of human .alpha.-galactosidase A is needed which can be incorporated into an expression vector under control of a strong promoter. Furthermore, this vector should provide stable expression of the cDNA and use a host system in which the processing and glycosylation may occur. Finally, biologically active enzyme must be produced. One such expression vector is provided by the baculovirus expression system of the present invention.
Baculoviruses infect Lepidopterm insects and have proven useful as recombinant expression vectors (Smith et al ., 1983, Mol. Cell. Biol. 3: 2156-2165; Luckow et al., 1988, Bio./Technology. 6: 47-55). The latter of these two references provides a detailed description of the available baculovirus vectors, methodology for their use and a list of proteins which have been expressed in this system. The particular advantage of baculovirus expression systems are very high levels of production (1 mg to 500 mg of protein per liter of culture have been reported), glycosylation and processing of the so-produced proteins. The early baculovirus expression vectors employ a strong promoter for a nonessential gene, the polyhedrin gene. To facilitate cloning, a DNA sequence encoding several restriction endonuclease sites had been inserted into the polyhedrin promoter (Luckow, et al., 1988, Bio/Technology 6: 47-55). Subsequently, it was discovered that this genetically engineered promoter was less effective (2-1000 fold) than the wild type polyhedrin promoter (Page, 1987, Nuc. Acids Res. 17 454; Ooi et al., 1989, J. Molec. Biol. 210 721-736). The present invention thus employs both the genetically altered polyhedrin promoter and the wild type polyhedrin promoter to generate expression systems providing large amounts of active .alpha.-galactosidase A.